A cellular framework for structural analysis and optimization


Canyurt O. E. , Hajela P.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, vol.194, pp.3516-3534, 2005 (Peer-Reviewed Journal) identifier identifier

  • Publication Type: Article / Article
  • Volume: 194
  • Publication Date: 2005
  • Doi Number: 10.1016/j.cma.2005.01.014
  • Journal Name: COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
  • Journal Indexes: Science Citation Index Expanded, Scopus
  • Page Numbers: pp.3516-3534
  • Keywords: structural analysis and designs, cellular automata, cellular genetic algorithm, AUTOMATA

Abstract

The present paper describes the implementation of it cellular automata based computational approach in structural analysis and design. This approach assumes that a computational domain can be subdivided into a number of discrete cells, with state variables associated with each cell. Collectively, these cell states define the state of the entire domain, and which may be evolved through application of local rules of interaction that apply to a defined neighborhood around each cell. The computational process is intrinsically parallel in nature and allows for a natural implementation on parallel computers. The paper describes an overview of this computational model for (he analysis of structural systems, and demonstrates how it can be extended for a fine-grained parallel implementation of the genetic algorithm based optimization strategy. The paper also illustrates an extension of the proposed model in solving problems of simultaneous analysis and optimization. (c) 2005 Published by Elsevier B.V.